© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 23910
    Prev
    Next
    INDEX
    Explanations

    prepositions and their corresponding phrases that indicate transformation or conversion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    176
    +0.14
    0.8%
    205
    +0.13
    0.7%
    358
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.14
    0.03
    142
    +0.13
    0.05
    356
    +0.13
    0.05
    Negative Logits
    ł
    -1.93
    stein
    -1.90
     $[
    -1.69
     myself
    -1.63
    osal
    -1.60
    MENTS
    -1.55
    ESULT
    -1.51
    ij
    -1.48
    Ļª
    -1.48
    ¦
    -1.47
    POSITIVE LOGITS
     inspiration
    1.70
     fame
    1.67
     vitality
    1.53
     assistant
    1.52
     convenience
    1.51
     excellence
    1.48
     offering
    1.46
     vibrant
    1.44
     thrive
    1.43
     prosperity
    1.42
    Activations Density 0.400%

    No Known Activations